DESCRIPTION: We are interested in the neuronal mechanisms by which the nervous system generates and coordinates rhythmic movements. Rhythmically active neural networks - central pattern generators (CPGs) -program the underlying motor pattern for these movements. In invertebrates - where the restricted number, large size, and identifiability of neurons offer technical advantages - progress in understanding CPGs and their adaptive modulation has been rapid. The study of these networks contributes not only to our understanding of motor systems but to our general understanding of the dynamics of neuronal ensembles. We analyze at the cellular and network level the heartbeat CPG in the leech This CPG comprises a network of oscillatory heart interneurons that interact synaptically among themselves and with segmental heart motor neurons. We use models that incorporate biophysical detail about intrinsic currents and synapses as tools to generate experimentally testable hypotheses and to extract principles of organization that can be transferred to other networks. Our immediate goal is complete models of the heartbeat CPG and of its interaction with segmental heart motor neurons to produce the heartbeat motor pattern. Our proposal is a step-by-step process for achieving this goal. 1a. To determine the strength and dynamics of the inhibitory synapses from two newly discovered premotor heart interneuron pairs onto rear heart motor neurons, and their activity pattern. 1b. To determine the intrinsic properties of the heart motor neurons and the properties of their intrasegmental electrical coupling and to analyze how heart motor neurons respond to inhibitory input and electrical coupling using dynamic clamp. 2. To determine the strength and dynamics of inhibitory connections and electrical coupling among heart interneurons. 3a. To construct a model of the heart motor neuron ensemble. 3a. To construct a canonical model of the heartbeat CPG. The models of this proposal cannot be separated from the experiments;the models are partially constructed, and through their deficiencies and successes they have generated and will continue to generate testable hypotheses and suggested new measurements to constrain parameters. Through the close interaction of modeling and experiments, we will refine the models and identify mechanisms by which oscillatory premotor networks attain functional phase relations and how they interact with motor neurons to produce coordinated behavior. This process will provide insights generally applicable to the normal functioning of CPGs and their coordination of motor neurons that will be useful in understanding the reaction of these networks to injury and disease. PUBLIC HEALTH RELEVACE Rhythmically active neural networks - central pattern generators (CPGs) - program rhythmic movements such as breathing and locomotion. In invertebrates - where the restricted number, large size, and identifiability of neurons offer technical advantages - progress in understanding CPGs has been rapid. The study of these networks will provide insights generally applicable to the normal functioning of CPGs that will be useful in understanding the reaction of CPG networks in vertebrates to spinal cord injury and disease and contribute to our general understanding of the dynamics of neuronal ensembles.